Published online before print March 22, 2007, doi: 10.1161/STROKEAHA.106.476804
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2007;38:1442.)
© 2007 American Heart Association, Inc.
Original Contributions |
Prevalence and Determinants of Subdiaphragmatic Visceral Infarction in Patients With Fatal Stroke
From Assistance-Publique Hôpitaux de Paris (H.A., C.D., P.G.S., J.-J.H., P.A.), the Department of Neuropathology Raymond Escourolle, La Salpêtrière Hospital (F.G.-R., A.J., C.D., J.-J.H., P.A.), the Department of Neurology and Stroke Centre (H.A., J.L., F.G.-R., A.J., P.A.), and Department of Cardiology, Bichat University Hospital, Paris, France (P.G.S.).
Correspondence to Professor Pierre Amarenco, Department of Neurology and Stroke Centre, Bichat University Hospital, 46 rue Henri Huchard, 75018 Paris, France. E-mail pierre.amarenco{at}bch.aphp.fr
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background and Purpose— Arterial thromboembolism is a common cause of both visceral and brain infarctions. Because the cause of brain infarction is unknown in up to 39% of patients, the discovery of subdiaphragmatic visceral infarction (SDVI) in this context is important, but its frequency is unknown. We therefore investigated the prevalence of SDVI in subjects who died from stroke. We also evaluated the yield of SDVI diagnosis for stroke subtyping.
Methods— We performed a case–control study using a series of 815 consecutive autopsies of patients who had died from a neurological disease, including 350 with stroke (260 infarcts and 90 hemorrhages). We systematically assessed the presence of renal, splenic, and mesenteric infarction (no case of spinal cord was recorded) and analyzed their determinants in patients with stroke. Patients with other neurological diseases served as the control group.
Results— Renal infarction was the most frequent SDVI (10.2%), whereas mesenteric infarction was rare (1.1%). At least one SDVI was found in 16.9% of patients with stroke (38.7% of patients with a cardioembolic stroke) and in 5.1% of patients with other neurological diseases (adjusted OR=2.12; 95% CI=1.08 to 4.16). Among patients with stroke, a significant heterogeneity in the prevalence of SDVI was found across etiological stroke subgroups with only three patients (3.3%) with hemorrhagic stroke having an SDVI (2 mesenteric and one renal infarction) compared with 56 patients (21.5%) with ischemic stroke (P<0.0001). Among patients with brain infarction and a SDVI, 76.8% had a definite cardiac source of embolism.
Conclusions— In patients with fatal brain infarction, the prevalence of SDVI is higher than previously thought, especially in those with stroke attributed to cardiac emboli. Seeking SDVI may assist in the etiologic diagnosis of brain infarction
Key Words: autopsy • embolic stroke • embolism • stroke • visceral infarction
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The specific cause of cerebral infarction remains undetermined in a large number of patients.1 Although a thromboembolic mechanism is frequently suspected, it is often difficult to identify the source of embolism.1 Cardiac and aortic sources of embolic stroke may also cause renal infarction, mesenteric ischemia, and acute limb ischemia.2,3 Most of the epidemiology of subdiaphragmatic visceral infarction (SDVI) in patients with stroke is based on information derived from case reports on patients with cardiovascular disease. Furthermore, SDVI may be underdiagnosed in patients with stroke because it may be asymptomatic or the symptoms and complaints may be more difficult to identify in the context of stroke.4,5 Hence, the actual prevalence of SDVI in ischemic stroke is unknown. Improvements in CT imaging have resulted in a marked improvement in the diagnosis of SDVI and of related vascular occlusion.6 Such a diagnosis in patients with ischemic stroke may have an adjunctive role to clarify stroke mechanisms when an embolic mechanism is suspected. In the context of brain infarction, the presence of an SDVI mainly suggests a cardiac source of embolism or multiple arterial dissections7 or a thrombotic disease (eg, disseminated intravascular coagulation or thrombocytopenia), but no series have quantified the frequency of SDVI in patients with stroke.
We investigated the prevalence of SDVI in patients with fatal stroke and compared it with that in patients with other fatal neurological diseases. We then evaluated the yield of an SDVI diagnosis to classify stroke into etiologic subtypes. For this purpose, we used a large consecutive autopsy series of patients who died from stroke and other neurological diseases.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Data Source and Population
Data were derived from the Multiple Atherosclerosis Site in Stroke study, an autopsy database of 886 patients with neurological diseases performed at La Salpêtrière Hospital in Paris between November 1982 and February 1989, when the autopsy rate was 73%.8
Of these, 505 were patients with neurological diseases other than stroke (the control group) and 381 were patients with stroke, including 83 with brain hemorrhage, 288 with brain infarction, and 10 with both brain hemorrhage and infarction. Because incomplete clinical data (n=41) and/or missing autopsy examination of SVDI (n=54), 71 patients were excluded from this study (Figure 1). The proportion of patients excluded (n=71) was similar in both study groups (8.1% versus 7.9%, P=0.91).
|
Clinical history, risk factors, imaging data, premortem clinical diagnosis, and autopsy findings were extracted from patients’ medical records. The etiology of ischemic stroke was categorized according to clinical data and pathological description following the GENIC (Etude du Profil Génétique de l’Infarctus Cérébral) classification9: lacunar infarct, atherothrombotic, cardioembolic, rare causes (such as disseminated intravascular coagulation and other hematologic causes, vasculitis, or dissection), coexisting causes (2 or more possible etiologies as defined previously), and unknown causes (when no identifiable cause was found). Patients with aortic arch plaques or ascending thoracic aorta plaques were included in the group atherothrombotic.
This classification was done by adjudicators blind to information regarding SDVI.
Autopsy Study
Autopsies were performed according to a standardized protocol10 and to French regulations. Briefly, each autopsy report included a macroscopic assessment of all internal organs and of the brain, a detailed anatomy of the cerebral arteries (extra- and intracranial artery), the renal arteries, and the aorta with a systematic drawing of the site and extent of atherosclerosis. A significant stenosis of the renal artery was defined as more than 50% narrowing in the luminal. Cerebral artery stenoses were graded, blind to clinical subtypes, according to percent reduction in luminal diameter and were then categorized as 0, 1 (stenosis 30% to 74%), 2 (stenosis 75% to 99%), or 3 (occlusion) at any segment (these cutoffs were defined in 1963). The microscopic study always included the brain and the SDVI.
Statistical Analysis
Patients were divided into 2 groups according to the pathological evidence of stroke. We used Student t test to compare continuous variables and the 
2 test to compare proportions; Fisher exact test was used when the expected cell frequency was less than 5. Comparisons of the prevalence of SDVI between the 2 groups were adjusted for age, gender, and clinical risk factors using logistic regression analysis. Crude and adjusted ORs of stroke associated with presence of SDVI were calculated with their 95% CIs.
We also compared the prevalence of SDVI between the brain infarction and hemorrhage and between the ischemic stroke subtypes using Fisher exact test. Because there was significant heterogeneity in the prevalence of SDVI across ischemic stroke subtypes attributed to cardioembolic causes, we computed the age- and gender-adjusted ORs of cardioemboli pathology associated with type and number of SDVIs.
In post hoc analysis, we studied the clinical and autopsy characteristics associated with the presence of SDVI in patients with brain infarction. Statistical testing was done at the 2-tailed 
level of 0.05. Data were analyzed using the SAS package, release 9.1 (SAS Inst).
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Among 815 consecutive autopsies, 350 patients had pathologic evidence of stroke (median time between event and death, 13 days [interquartile range, 5 to 32 days]) and 465 patients had other neurological diseases (Figure 1). Patients with stroke were older, had more cardiovascular risk factors, and a more frequent history of cardiovascular events than patients with other neurological diseases (Table 1). The gender ratio did not differ between the 2 groups.
|
Autopsy Prevalence of Subdiaphragmatic Visceral Infarction
At least one SDVI was found in 16.9% (95% CI=12.9 to 20.8) of patients with stroke and in 5.1% (95% CI=3.1 to 7.2) of patients with other neurological diseases (crude OR=3.73; 95% CI=2.27 to 6.12; P<0.001). This difference persisted after adjustment for age, gender, and clinical risk factors (adjusted OR=2.12; 95% CI=1.08 to 4.16). Renal infarction was the most frequent SDVI in this autopsy study and occurred in 83 patients (10.2%), whereas mesenteric infarction was rare (1.1%) (Table 2). The adjusted ORs for renal and splenic infarction were 2.12 (95% CI=0.93 to 4.81, P=0.07) and 3.04 (95% CI=1.03 to 8.98, P=0.04), respectively. Twelve patients with stroke (3.4%) and one patient with another neurological disease (0.2%) had both renal and splenic infarction. None of the patients had mesenteric infarction associated with other types of SDVI.
|
Among patients with stroke, a significant heterogeneity in the prevalence of SDVI was found across etiologic stroke subgroups. SDVI was found in only 3 patients (3.3%) with hemorrhagic stroke (2 mesenteric infarctions and one renal infarction) compared with 56 patients (21.5%) with ischemic stroke (P<0.0001). Moreover, when ischemic stroke was divided into etiologic subtypes, SDVI was strongly associated with cardioembolic causes (Table 2). Of the 56 patients with ischemic stroke with SDVI, 64.3% were classified as the cardioembolic etiologic subtype and 12.5% were classified as having stroke with coexisting causes (combining atherothrombotic and cardioembolic causes). Among the 260 patients with ischemic stroke, the age- and gender-adjusted OR of cardiac source of embolism associated with the presence of any SDVI was 5.56 (95% CI=2.76 to 11.20). The adjusted OR of cardioembolic pathology was higher when both renal and splenic infarctions were present (Figure 2). Among the 12 patients with ischemic stroke with both renal and splenic infarctions, 10 had a definite cardiac source of embolism.
|
Subdiaphragmatic Visceral Infarction Among Patients With Brain Infarction
We studied the clinical and anatomic characteristics associated with the presence of any SDVI among the 260 patients with stroke. In univariable analysis, patients with SDVI had a higher heart weight than patients without SDVI (P=0.002; Table 3). Patients with SDVI were also more likely to have diabetes mellitus and a history of myocardial infarction.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The main finding of this study is that SDVI was more frequently observed in patients with stroke attributable to a definite cardiac source of embolism compared with others. This result was of course expected based on previous clinical data,11 but to the best of our knowledge, we did not find in the literature such a large autopsy series quantifying the frequency of SDVI in patients with stroke. Of 260 patients with brain infarction, 56 patients (21.5%) had at least one SDVI. Among them, 64.3% were classified as the cardioembolic etiologic subtype and 12.5% were classified as the etiologic subtype with coexisting causes (both atherothrombotic and cardioembolic). These autopsy results suggest that more than 75% of patients with ischemic stroke with SDVI had a definite cardiac source of embolism. In a study of 193 patients with ischemic stroke, Ramirez-Lassepas et al12 observed 3 patients with a history of systemic embolization; these patients had a cardiac source of embolism. All these findings suggest that, in patients with stroke, SDVI is frequently related to a cardiogenic mechanism.
Embolism of cardiac origin accounts for approximately 15% to 25% of ischemic strokes.13 With the growing use of transesophageal echocardiography, cardioaortic embolism can now be detected more frequently.14 However, in up to 39% of patients with brain infarction, the cause remained undetermined1; and in 13% of cases, a potential cardiac source of emboli coexists with a severe atherosclerotic stenosis of a cerebral artery or evidence of a small-vessel disease, making the diagnosis presumptive.9 Furthermore, in several cases, only a suggestive clinical–neuroimaging pattern suggests a cardiac embolism as the explanation for the brain infarction,15 but complete cardiac investigations do not confirm the diagnosis. These patients are usually classified into a single category ("stroke of unknown cause") and are frequently managed by aspirin, whereas systemic anticoagulation is preferable in the case of a cardiac source of embolism.16,17 For example, paroxysmal atrial fibrillation is an important risk factor for cerebral embolism18 and may occur in the absence of any demonstrable heart disease, which makes its diagnosis difficult. Because cardioembolism in our series and in others is the main cause of systemic embolization,2,11,19 the discovery of SDVI associated with cryptogenic stroke should be taken into account in the diagnosis of probable cardiac embolism and therefore may assist in making decisions for anticoagulant therapy.
Earlier studies20,21 described the association between complex atherosclerotic plaques in the thoracic aorta and peripheral embolic events. We did not find any association between SDVI and ulcerated plaques in the abdominal or thoracic aorta among patients with brain infarction, but we could not exclude that this association was overlooked as a result of the lack of statistical power.
Renal infarction was the most frequent SDVI in our series. Of 56 patients with ischemic stroke with SDVI, 42 (75%) had a renal infarction. Of these, 33 patients (79%) had an ischemic stroke with a potential cardiac source of embolism. Although renal infarction is usually caused by embolism or renal artery lesion,2,19,22,23 no relation was found with renal artery stenosis in our autopsy data.
Study Limitations
There are some limitations in this study. First, autopsy studies are not representative of the whole spectrum of stroke; therefore, our results apply only to patients with fatal stroke. Second, this cohort was collected in the 1980s. However, although treatment and outcomes have evolved largely since then, the pathophysiology, epidemiology, and associations between the various conditions are likely to remain stable. Therefore, there is no reason to expect that an association observed between SDVI and some cardiac abnormalities then would not hold true today. Finally, we cannot also exclude that the retrospective review of cardiovascular risk factors from the medical charts leads to some bias and underestimation of the prevalence of cardiovascular risk factors in this population. Today, because of a dramatic decline in postmortem studies, large autopsy studies are very rare. For this reason, this large autopsy series provides unique information.
Conclusions
SDVI was found in 22% of autopsied patients with brain infarction. We found a highly significant association between SDVI and cardioembolic causes. Renal infarction was the most frequent among all SDVI. These results can assist in decision-making in patients with cryptogenic brain infarction. Based on these results, we believe that the yield of an abdominal CT scan should now be evaluated in the workup of patients with brain infarction of unknown cause and a clinical or radiological thromboembolic pattern. Follow-up studies are required to extend our results to nonfatal strokes and to assess the clinical use of SDVI diagnosis in patients with ischemic stroke.
|
Acknowledgments |
|---|
Sophie Rushton-Smith, PhD, provided editorial assistance in the preparation of this manuscript and was funded by SOS-ATTAQUE CEREBRALE Association.
Disclosures
None.
Received October 29, 2006; accepted December 6, 2006.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Sacco RL, Ellenberg JH, Mohr JP, Tatemichi TK, Hier DB, Price TR, Wolf PA. Infarcts of undetermined cause: The NINCDS stroke data bank. Ann Neurol. 1989; 25: 382–390.[CrossRef][Medline] [Order article via Infotrieve]
2. Argiris A. Splenic and renal infarctions complicating atrial fibrillation. Mt Sinai J Med. 1997; 64: 342–349.[Medline] [Order article via Infotrieve]
3. Bradbury AW, Brittenden J, McBride K, Ruckley CV. Mesenteric ischaemia: a multidisciplinary approach. Br J Surg. 1995; 82: 1446–1459.[Medline] [Order article via Infotrieve]
4. Park WM, Gloviczki P, Cherry KJ Jr, Hallett JW Jr, Bower TC, Panneton JM, Schleck C, Ilstrup D, Harmsen WS, Noel AA. Contemporary management of acute mesenteric ischemia: factors associated with survival. J Vasc Surg. 2002; 35: 445–452.[CrossRef][Medline] [Order article via Infotrieve]
5. Iga K, Izumi C, Nakano A, Sakanoue Y, Kitaguchi S, Himura Y, Gen H, Konishi T. Problems in the initial diagnosis of renal infarction. Intern Med. 1997; 36: 330–332.[Medline] [Order article via Infotrieve]
6. Ishikawa I, Masuzaki S, Saito T, Yuri T, Shinoda A, Tsujigiwa M. Magnetic resonance imaging in renal infarction and ischemia. Nephron. 1989; 51: 99–102.[Medline] [Order article via Infotrieve]
7. Amarenco P, Seux-Levieil ML, Cohen A, Levy C, Touboul PJ, Bousser MG. Carotid artery dissection with renal infarcts: two cases. Stroke. 1994; 25: 2488–2491.[Abstract]
8. Amarenco P, Duyckaerts C, Tzourio C, Henin D, Bousser MG, Hauw JJ. The prevalence of ulcerated plaques in the aortic arch in patients with stroke. N Engl J Med. 1992; 326: 221–225.[Abstract]
9. Touboul PJ, Elbaz A, Koller C, Lucas C, Adrai V, Chedru F, Amarenco P. Common carotid artery intima-media thickness and brain infarction: The etude du profil genetique de l’infarctus cerebral (GENIC) case– control study. The GENIC investigators. Circulation. 2000; 102: 313–318.
10. Jaramillo A, Gongora-Rivera F, Labreuche J, Hauw JJ, Amarenco P. Predictors for malignant middle cerebral artery infarctions: a postmortem analysis. Neurology. 2006; 66: 815–820.
11. Kittner SJ, Sharkness CM, Price TR, Plotnick GD, Dambrosia JM, Wolf PA, Mohr JP, Hier DB, Kase CS, Tuhrim S. Infarcts with a cardiac source of embolism in the NINCDS stroke data bank: historical features. Neurology. 1990; 40: 281–284.
12. Ramirez-Lassepas M, Cipolle RJ, Bjork RJ, Kowitz J, Snyder BD, Weber JC, Stein SD. Can embolic stroke be diagnosed on the basis of neurologic clinical criteria? Arch Neurol. 1987; 44: 87–89.
13. Cardiogenic brain embolism. The second report of the Cerebral Embolism Task Force. Arch Neurol. 1989; 46: 727–743.
14. Cujec B, Polasek P, Voll C, Shuaib A. Transesophageal echocardiography in the detection of potential cardiac source of embolism in stroke patients. Stroke. 1991; 22: 727–733.
15. Ferro JM. Cardioembolic stroke: an update. Lancet Neurol. 2003; 2: 177–188.[CrossRef][Medline] [Order article via Infotrieve]
16. Stroke Prevention in Atrial Fibrillation Study. Final results. Circulation. 1991; 84: 527–539.
17. Risk factors for stroke and efficacy of antithrombotic therapy in atrial fibrillation. Analysis of pooled data from five randomized controlled trials. Arch Intern Med. 1994; 154: 1449–1457.
18. Lin HJ, Wolf PA, Benjamin EJ, Belanger AJ, D’Agostino RB. Newly diagnosed atrial fibrillation and acute stroke: The Framingham Study. Stroke. 1995; 26: 1527–1530.
19. Mahapatra RK, Agarwal JB, Chopra P. Systemic thromboembolism in rheumatic heart disease. Jpn Heart J. 1980; 21: 773–777.[Medline] [Order article via Infotrieve]
20. Khatibzadeh M, Mitusch R, Stierle U, Gromoll B, Sheikhzadeh A. Aortic atherosclerotic plaques as a source of systemic embolism. J Am Coll Cardiol. 1996; 27: 664–669.[Abstract]
21. Tunick PA, Perez JL, Kronzon I. Protruding atheromas in the thoracic aorta and systemic embolization. Ann Intern Med. 1991; 115: 423–427.
22. Ramamoorthy SL, Vasquez JC, Taft PM, McGinn RF, Hye RJ. Nonoperative management of acute spontaneous renal artery dissection. Ann Vasc Surg. 2002; 16: 157–162.[CrossRef][Medline] [Order article via Infotrieve]
23. Hazanov N, Somin M, Attali M, Beilinson N, Thaler M, Mouallem M, Maor Y, Zaks N, Malnick S. Acute renal embolism: forty-four cases of renal infarction in patients with atrial fibrillation. Medicine (Baltimore). 2004; 83: 292–299.[CrossRef][Medline] [Order article via Infotrieve]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||